Dat/Hyperdrive Performance Monitor

[joehand]

This project will test & monitor the performance the Dat/Hyperdrive Stack.

1. Hypercore (hypercore-stats-ui)
2. Peer Discovery
3. Dat CLI Share/Download (dat-performance)

   Who are the users?

Dat/Hyperdrive Developers

How would the user interact with it?

1. Graphs of performance over time for a few file sets (example)
2. View performance results via CLI or Browser on local computer

   Why is the project important to the ecosystem?

It is essential for us to be able to improve performance over time and see any regressions. It'd be great to do this automatically, without having to do it as we work on development.

What defines the minimum requirements to sufficiently release (version 1)

• Visualize hypercore performance (hypercore-stats-ui)
• Peer discovery visualization & performance
• Dat CLI share/download performance for several types of Dats

  What are some stretch goals or interesting features for further releases (version 2, 3)

• Visualize performance
• Track performance over time
• CI testing, automatic tests for branches/PRs

[max-mapper]

Here's what I think the 'Peer discovery visualization & performance' portion should include (for browser clients):

• Which signalhubs did you talk to?
• How long did it take to talk to each signalhub?
• How many peers did you get back from each signalhub?
• For each peer, what is the status of establishing a RTCPeerConnection?
• Did the connection fail or succeed? Did it use ICE/STUN or TURN?
• How much data is flowing through each connection?